1.1.2015-31.12.2019 Aalto University
Professor Paavo Alku (Aalto University) is one of the world’s leading researchers in his field. He is involved in an interdisciplinary research project aimed at developing new methods of statistical speech synthesis. His particular research interest is the role of the vocoder at the heart of statistical speech synthesis: the aim is to make use of signal processing methods that more accurately model the physiology of human speech production. To this end the project will be using a computational inverse method.
Professor Alku and his team are especially interested in creating natural sounding synthesis of more challenging types of speech, such as child’s speech or emotional speech. The results of their research will have extensive application in the ICT sector, particularly in developing more natural speech interfaces. The methods will be integrated with a synthetic system that will be able to generate speech from any system. The results have potential application in speech-to-speech translation, in the gaming industry, in language teaching and in assisting people with speech disorders.
Professor Alku and his team have for years been doing pioneering work to develop methods in statistical parametric speech synthesis. During his term as Academy Professor, Alku will take research in this field to yet another level.